In-depth analysis on Credit Writedowns Pro.

Reforming the banks

By Michael Pettis

I just got back from a very interesting but hectic week in New York and Washington, followed by two days at a conference in Hangzhou. During my meetings I noticed that much of the discussion, and many of the questions I was asked by both government officials and investors, focused on debt levels and reforms in the Chinese financial system. I have written a lot about rising debt in China and am glad that analysts and policymakers seem to be spending a lot more time thinking about balance sheet issues. Every case of rapid, investment-driven growth in the past century, as far as I can make out, has at some point reached a stage in which debt levels rose to unsustainable levels and precipitated either a debt crisis or a long grinding adjustment period.

The reason debt levels always seem to grow unsustainably, I suspect, is that in the initial stages of the growth model much if not all of the investment is economically viable as it pours into building necessary infrastructure whose profits and externalities exceed the cost of the investment. The result is real growth. At some point, however, the combination of subsidies, distorted incentives (in which investment benefits accrue to those making the investment while costs are shared broadly through the banking system), and very cheap financing costs leads inexorably to wasted investment and debt rising faster than asset values. This is when the debt burden begins to rise in an unsustainable way.

By that point, however, the system is so addicted to investment-driven growth that it is not able easily to reverse or unwind the process until it is too late and debt levels have become a significant problem. Look at the economic “miracles” of the past fifty years – the Soviet Union in the 1950s and 1960s, parts of Latin American and especially Brazil in the 1960s and 1970s, Japan in the 1970s and 1980s, the Asian Tigers in the 1980s and 1990s.

All of them experienced astonishing investment-driven growth for many years, followed by unsustainably rising debt levels and either crises or “lost decades”. On that note I should mention that on the plane back I re-read Jeffrey Frieden’s excellent Debt, Development and Democracy, about Latin America from 1965 to 1985, and it provides a great explanation on how the proceeds of rising debt are distributed.

What do banks do?

Whether or not we have reached the point in China in which investment is misallocated and debt levels rising is clearly a matter for heated debate – I think we have already passed that point – but clearly we are tending in that direction. The key problem, I think, is the way in which the financial system allocates capital. Every financial system is capable of periods of capital misallocation, and this almost always seems to happen during periods of very low interest rates and rapid money expansion, but some financial systems do this more extravagantly than others.

So why do some financial systems misallocate capital this more than others? As I see it there are broadly speaking two very different conceptions of the role of a country’s financial systems. In one, banks act largely as fiscal agents for the government or the economic elite, accumulating savings and deploying capital into projects usually selected for promotion by those elites. Typically the key objectives in this kind of banking system are rapid elite-directed growth and overall financial stability.

Since banks are in the business of taking risk, and since rapid credit expansion is inherently risky, the only way to guarantee financial stability is to extract much or all the risk from the banks and imbed them elsewhere. In practice the only “elsewhere” big enough is the state. In this kind of banking system the state typically socializes credit risk and passes losses onto taxpayers or depositors.

France’s Société Générale du Crédit Mobilier, established in 1852, is in my opinion one of the pioneers of this conception of banking, although of course state-directed banks are much older than that. The history of banks like the Bank of England and John Law’s Mississippi Company shows how closely intertwined banking and state objectives have been for a very long time.

These kinds of banking systems can generate tremendous economic growth, at least for countries that are economically and technologically undeveloped and in which it is relatively easy to identify projects that generate economic value. It may be much harder to identify obviously good projects, however, in more advanced countries, or for countries whose infrastructure is well developed for their levels of wealth and worker productivity.

In that case these kinds of financial systems inevitably run into the problem of capital misallocation. It doesn’t matter if at one point they do a great job of allocating capital and generating real growth. As long as the same allocation process is maintained, it seems, at some point they begin to overinvest. Perhaps this is because the economic sectors that benefit most from the regulatory, credit and economic subsidies, not surprisingly, become increasingly powerful within the political system and increasingly reluctant to allow the system to change. Whatever the reason, this is the kind of financial system, I would argue, that has a built-in tendency eventually to misallocate capital more extravagantly.

The other type of system, in which the problem of systematic capital misallocation is much reduced, is one in which banks decide for themselves the kinds of activities they fund, and their shareholders and depositors bear both the rewards and risks of their capital allocation. These kinds of banking system are much more prone to instability, but they are also much more efficient at allocating capital over the long term. In part this is because there is a fairly robust mechanism for recognizing and liquidating poor investment. In the former system, because risk tends to be socialized, there is no obvious mechanism, besides that of an omniscient and disinterested credit committee, for identifying and correcting misallocation.

In the latter system investors who have to bear the risk are responsible for monitoring the risk and forcing liquidation. Today we are likely to describe this as an “Anglo-Saxon” system, but it just as easily characterizes private banking in France during the 19th century and even, perhaps ironically, banking in 19th century China (the piahao banks from Shanxi province, for example) and the first half of the 20th century, with their many private and informal banks.

The recent global financial crisis has seriously undermined the prestige of the “Anglo-Saxon” model, but we need to be a little careful about throwing the baby out with the bathwater. For all the absurdity in real estate lending (which to me was more likely to have been caused by excessively loose monetary policy than by flaws inherent to the system), the financial institutions as a whole have done a pretty good job in allocating capital productively over the long term and, for example, were instrumental in funding the various technological booms we’ve had in recent decades. In fact I suspect that the prestige of the Anglo-Saxon model soared in the past two decades precisely because its biggest competitor for prestige, the Japanese banking system, collapsed so spectacularly in the 1990s.

In practice of course there is no pure example of one financial system or the other, but as the statement above suggests it is pretty safe to say that Japan during its growth period, and the countries that copied the Japanese model, are closet to the extreme version of the former. The current Chinese financial system, even more than Japan, is clearly one in which the purpose of the financial system is to act as the state’s fiscal agent and in which banking stability is guaranteed by the state. It is also clearly one in which capital misallocation can become a huge problem.

Has there been reform?

It is in this context that we need to understand what it means to refer to banking or financial sector reform in China. Last Friday I spoke in Washington at a conference, organized by the Carnegie Endowment, on China’s economic prospects in the next five years and the subject came up. Peter Botelier was one of the other members of the panel and during our presentations the issue of banking reform was brought up. We agreed fundamentally on a lot of things but he was more optimistic than I was about whether or not there had been real financial sector reform in the past decade. He thought there had been, and mentioned the IPOs, the creation of modern credit committees, and a number of other things.

I argued that there had been very limited meaningful reform. As I see it, banking or financial sector reform in China is meaningful only to the extent that it shifts China’s banking system from the first of the two systems described above closer to the latter. This is not to say that it must go from one extreme to the other – only that it must move in the direction of the other.

Why? Because much of China’s most obvious investment has been identified and funded over the past three decades, and in the last ten years the combination of socialized credit risk, very low interest rates, state-directed lending and tremendous pressure on the part of SOEs and local and municipal governments to generate employment and growth in the short term has increased the probability that the Chinese financial system may be misallocating capital on a dangerous scale. The growth in bank assets, in other words, would be less than the growth in bank liabilities if both were correctly valued as a function of discounted expected cash flows.

Why am I so sure? Aside from the many studies I’ve cited showing that profitability in many of China’s largest companies is substantially less than the value of the financing and other subsidies, and anecdotal evidence of unnecessary real estate and infrastructure projects, just imagine what would happen to banking deposits and stock prices if the government credibly removed all guarantees on loans extended by the banks, and furthermore removed interest rate controls. I suspect most investors and depositors would assume, correctly in my opinion, a surge in non-performing loans that would wipe out the banks’ capital base, and so would sell their stocks and withdraw their deposits.

The fact that this is unlikely to happen is irrelevant. It just means that the losses are hidden and transferred to the state, and via the state, to households. If that is the case, then since the banking system can no longer easily identify economically viable projects and is in fact wasting money, the usefulness of the bank-as-fiscal-agent model is much reduced. We need now to have banks in China that can correctly identify economically useful projects in which to invest and limit their credit growth to those projects.

This is, I think, pretty clearly the attitude of financial regulators at the PBoC and the CBRC. They are concerned about the pace of credit growth, which would not be a problem at all if credit were going to economically viable projects. After all, I would guess that the only significant systemic risks that banks take on are credit risk and maturity mismatch, and Chinese banks don’t have to worry about the latter (no bank runs).

If we agree that reform in the Chinese context means moving away from the fiscal-agent model and towards one with stronger internal incentives for monitoring capital allocation, then most Chinese economists would probably agree that in the past two years reform has gone backward. There is however also a view among many academics – one that I share – that there has been very little meaningful reform at all, at least in the past decade.

What is reform?

That may seem like a strange thing to say, especially since many analysts, especially bank research analysts, have lauded the significant reforms the Chinese financial system has undergone in the past decade. To me however these reforms – the introduction of QFIIs and later QDIIs, the growth of derivatives, bank IPOs, etc. – are largely beside the point. As I see it financial reform in China really means four things, none of which have been seriously implemented:

  1. Interest rates must be liberalized so that the true cost of capital is reflected in evaluating the worth of a project. All central banks intervene in interest rates, if only to smooth out seasonal and temporary volatility, but PBoC artificially sets the rates for all maturities at least 400-800 basis points too low. By keeping the cost of capital so low, it disguises the true cost to China of capital and permits investment in projects whose returns are simply not justified.
  2. Corporate governance must be reformed, and this means in part a significant reduction in the number of projects whose risks are socialized. Borrowers and banks must act on economic rather than non-economic issues, and as long as risk is socialized – implicitly or explicitly – there is no need to worry about the riskiness of repayment prospects. Remember how a much milder socialization of credit risk, the so-called “Greenspan put”, distorted lending and investment decisions in the US.
  3. The regulatory framework must be stabilized and government intervention should become much more predictable, at least on economic grounds. Investors should be in the business of predicting what economical value will be created, not what steps the government will take next.
  4. Information quality must be sharply improved – macroeconomic information as well as financial statements. It is pointless to ask investors to make decisions about the future if they have poor or systematically biased information with which to work.

I would argue that any “reform” introduced into the financial system is ineffective as reform if it does not materially affect one of the above four. So has there been real financial reform?

To take the last point first, I would argue that the National Bureau of Statistics and the People’s Bank of China have done great jobs in improving the quality of macroeconomic and financial sector data, but there still is a long way to go, especially in the quality of financial statements. In that sense, there has been some real reform of the banking and financial systems in the past decade.

On the other three matters, however, I would argue that there has been very little change at all, expect maybe some backward movement in corporate governance in the past three years. There is from time to time some talk about eventually liberalizing interest rates, but interest rates are as controlled as they have ever been (in fact real rates have declined in the past several months to seriously negative rates) and I don’t think anyone expects anything to happen soon on that front.

Banks compete heavily for deposits, but they cannot compete on price, and any attempt to get around the system – for example when banks offer gifts to attract deposits – is prohibited. Many would argue that the PBoC cannot liberalize interest rates now because if they did, and rates soared as they would be expected to do, we would see a surge in bankruptcies. This is true of course, but it is equally true that the longer we wait, the more difficult it becomes for exactly that reason.

Some have argued that bond and money market rates are set by the markets, so to the extent that these markets are growing we are seeing gradual liberalization of interest rates, but I think this argument is mistaken. Banks are the biggest buyer of these instruments and they are definitely the price-setters. Since the key issue for them is their own cost of funding and their lending alternatives, the PBoC largely determines prices in the bond and money markets via its setting of deposit and lending rates.

As for corporate governance reform, and removing implicit and explicit guarantees on risk, clearly neither has happened. Like in the case of interest-rate liberalization, there would be a heavy cost if this were done too quickly, but of course the more debt levels build up the heavier the cost.

So I think we need to be a little skeptical when we hear about the tremendous reforms that the financial system has undergone in the past decade, with the implication that things are going to continue to improve. I think in the 1990s there certainly were important reforms, but I would argue that if we are indeed at the point where capital is being misallocated in the aggregate, then meaningful reform requires movement on the above issues. To the extent that there hasn’t been any real movement, there has been no real reform.

This is an abbreviated version of the newsletter that went out last week. Academics, journalists, and government and NGO officials who want to subscribe to the newsletter should write to me at chinfinpettis@yahoo.com, stating your affiliation, please. Investors who are clients of Shenyin Wanguo Securities will already receive the newsletter. Investors who are not clients but who want to buy a subscription should write to me. also at that address.

Michael Pettis

About 

Michael Pettis is a Senior Associate at the Carnegie Endowment for International Peace and a finance professor at Peking University’s Guanghua School of Management, where he specializes in Chinese financial markets. Pettis has worked on Wall Street in trading, capital markets, and corporate finance since 1987. Pettis is a member of the Institute of Latin American Studies Advisory Board at Columbia University as well as the Dean’s Advisory Board at the School of Public and International Affairs. He received an MBA in Finance in 1984 and an MIA in Development Economics in 1981, both from Columbia University. He writes the blog .

1 Comment

  1. Anonymous says:

    One of the reasons why banks make so many bad loans is because they do not encourage original thinking from loan officers. There is far too much of copying what has been approved before. Far too much herd behaviour from banks. They cannot fire you for making loans that follow many others before. Even if it has created a bubble. That is the problem of directors who have abdicated responsibility of overall lending. So what happens is that new businesses are ignored and lending tied to collateral of some kind is common.

    In some respects banks create the mess. If a business wants to expand the banks demand collateral. If the owner has put all their collateral in they stand to lose everything. The bank will still want more collateral and this forces businesses to either give up and self fund or bring in other partners and collateral or lose out. This hampers new business growth. Also the big banks are happier making big loans to the big companies thinking that they will be more secure.

    At the height of the last bubble UK banks had more than 70% of lending secured against property. It was why they were so vulnerable to a property crash following the bubble.

    What would help banks is possibly banning them from the residential mortgage market. Leave this to savings and loans or mutually funded building societies. This would leave banks with unsecured loans and business loans. They would have to be a lot more careful, which might temper excessive credit growth. If banks had a limit to the central banks lender of last resort facilities it might also make them temper excessive loan growth. If a bank had to make a request for emergency funds it immediately starts a bank investigation by the regulator. During which all loans over $1 million are rationed. They could lose it business. An immediate 24 month ban on any directors bonus payout even deferring would banned. This would make the lender of last resort a serious last resort. Banks would do everything to avoid such a situation.

    Longer term the Basel agreements need to be super simple. Risk based models have clearly failed as they ignored long tail events such as black swan events and a simple maximum leverage ratio for classes of banks needs to be internationally agreed. Commercial banks with a maximum of 15 times capital, investment banks 3 times capital, trade banks 10 times capital. Credit cards 10 times capital. Shadow banks and all other non banks will be limited to 3 times capital. These would limit the overall risk for banks and financial markets and force non banks out of business or to cap their risks systemically. Any bank that has more than a set amount of business or is simply too big has to break itself up. This will create competition. Universal or mega banks will have to have capital requirements overall in line with their riskiest business. So the mega banks would have to have much higher capital base or hive off the investment banks business.

    Central banks could also do a lot more in terms of stopping bubbles. If a bank made too many loans to one sector creating a dominant position, or creating a bubble the central banks could demand that the bank hand over some capital. A reduction of $1 billion of capital would reduce $15 billion of loans and profits associated with that business. Alan Greenspan would claim that it is impossible to spot bubbles but many people can see them even if central bankers close their eyes to the problems.

    Ultimately fixed leverage ratios for banks will do a lot better than the current crop of central bankers as they simply do not regard regulation as part of their remit.

  2. DavidLazarusUK says:

    One of the reasons why banks make so many bad loans is because they do not encourage original thinking from loan officers. There is far too much of copying what has been approved before. Far too much herd behaviour from banks. They cannot fire you for making loans that follow many others before. Even if it has created a bubble. That is the problem of directors who have abdicated responsibility of overall lending. So what happens is that new businesses are ignored and lending tied to collateral of some kind is common.

    In some respects banks create the mess. If a business wants to expand the banks demand collateral. If the owner has put all their collateral in they stand to lose everything. The bank will still want more collateral and this forces businesses to either give up and self fund or bring in other partners and collateral or lose out. This hampers new business growth. Also the big banks are happier making big loans to the big companies thinking that they will be more secure.

    At the height of the last bubble UK banks had more than 70% of lending secured against property. It was why they were so vulnerable to a property crash following the bubble.

    What would help banks is possibly banning them from the residential mortgage market. Leave this to savings and loans or mutually funded building societies. This would leave banks with unsecured loans and business loans. They would have to be a lot more careful, which might temper excessive credit growth. If banks had a limit to the central banks lender of last resort facilities it might also make them temper excessive loan growth. If a bank had to make a request for emergency funds it immediately starts a bank investigation by the regulator. During which all loans over $1 million are rationed. They could lose it business. An immediate 24 month ban on any directors bonus payout even deferring would banned. This would make the lender of last resort a serious last resort. Banks would do everything to avoid such a situation.

    Longer term the Basel agreements need to be super simple. Risk based models have clearly failed as they ignored long tail events such as black swan events and a simple maximum leverage ratio for classes of banks needs to be internationally agreed. Commercial banks with a maximum of 15 times capital, investment banks 3 times capital, trade banks 10 times capital. Credit cards 10 times capital. Shadow banks and all other non banks will be limited to 3 times capital. These would limit the overall risk for banks and financial markets and force non banks out of business or to cap their risks systemically. Any bank that has more than a set amount of business or is simply too big has to break itself up. This will create competition. Universal or mega banks will have to have capital requirements overall in line with their riskiest business. So the mega banks would have to have much higher capital base or hive off the investment banks business.

    Central banks could also do a lot more in terms of stopping bubbles. If a bank made too many loans to one sector creating a dominant position, or creating a bubble the central banks could demand that the bank hand over some capital. A reduction of $1 billion of capital would reduce $15 billion of loans and profits associated with that business. Alan Greenspan would claim that it is impossible to spot bubbles but many people can see them even if central bankers close their eyes to the problems.

    Ultimately fixed leverage ratios for banks will do a lot better than the current crop of central bankers as they simply do not regard regulation as part of their remit.